Spring Batch 配置重试逻辑

Spring Batch在处理过程中遇到错误job默认会执行失败。为了提高应用程序的健壮性，我们需要处理临时异常造成失败。本文我们探讨如何配置Spring Batch的重试逻辑。

1. 应用示例

批处理应用读取csv文件：

sammy, 1234, 31/10/2015, 10000
john, 9999, 3/12/2015, 12321

然后，通过调用rest接口处理每条记录，获取用户的年龄和邮编属性，为了正确输出日期，可以在属性上增加@XmlJavaTypeAdapter(LocalDateTimeAdapter.class)注解：

@XmlRootElement(name = “transactionRecord”)
@Data
public class Transaction {
private String username;
private int userId;
private int age;
private String postCode;
private LocalDateTime transactionDate;
private double amount;
}

处理类如下：

public class RetryItemProcessor implements ItemProcessor<Transaction, Transaction> {

    private static final Logger LOGGER = LoggerFactory.getLogger(RetryItemProcessor.class);

    @Autowired
    private CloseableHttpClient closeableHttpClient;

    @Override
    public Transaction process(Transaction transaction) throws IOException, JSONException {
        LOGGER.info("Attempting to process user with id={}", transaction.getUserId());

        HttpResponse response = fetchMoreUserDetails(transaction.getUserId());
        //parse user's age and postCode from response and update transaction
        String result = EntityUtils.toString(response.getEntity());
        JSONObject userObject = new JSONObject(result);
        transaction.setAge(Integer.parseInt(userObject.getString("age")));
        transaction.setPostCode(userObject.getString("postCode"));

        return transaction;

    }

    private HttpResponse fetchMoreUserDetails(int id) throws IOException {
        final HttpGet request = new HttpGet("http://www.testapi.com:81/user/" + id);

        return closeableHttpClient.execute(request);
    }

}

这里当然也可以使用RestTemplate进行调用，调用服务仅为了测试，读者可以搭建测试接口。
最终输出结果为：

<transactionRecord>
    <transactionRecord>
        <amount>10000.0</amount>
        <transactionDate>2015-10-31 00:00:00</transactionDate>
        <userId>1234</userId>
        <username>sammy</username>
        <age>10</age>
        <postCode>430222</postCode>
    </transactionRecord>
    ...
</transactionRecord>

2. 给处理增加重试功能

如果连接rest接口因为网络不稳定导致连接超时，那么批处理将失败。但这种错误并不是不能恢复，可以通过重试几次进行尝试。因此我们配置批处理job在失败的情况下重试三次。

@Configuration
@EnableBatchProcessing
public class SpringBatchRetryConfig {
    private static final String[] tokens = { "username", "userid", "transactiondate", "amount" };
    private static final int TWO_SECONDS = 2000;

    @Autowired
    private JobBuilderFactory jobBuilderFactory;

    @Autowired
    private StepBuilderFactory stepBuilderFactory;

    @Value("input/recordRetry.csv")
    private Resource inputCsv;

    @Value("file:xml/retryOutput.xml")
    private Resource outputXml;

    public ItemReader<Transaction> itemReader(Resource inputData) throws ParseException {
        DelimitedLineTokenizer tokenizer = new DelimitedLineTokenizer();
        tokenizer.setNames(tokens);
        DefaultLineMapper<Transaction> lineMapper = new DefaultLineMapper<>();
        lineMapper.setLineTokenizer(tokenizer);
        lineMapper.setFieldSetMapper(new RecordFieldSetMapper());
        FlatFileItemReader<Transaction> reader = new FlatFileItemReader<>();
        reader.setResource(inputData);
        reader.setLinesToSkip(1);
        reader.setLineMapper(lineMapper);

        return reader;
    }

    @Bean
    public CloseableHttpClient closeableHttpClient() {
        final RequestConfig config = RequestConfig.custom()
          .setConnectTimeout(TWO_SECONDS)
          .build();

        return HttpClientBuilder.create().setDefaultRequestConfig(config).build();
    }

    @Bean
    public ItemProcessor<Transaction, Transaction> retryItemProcessor() {

        return new RetryItemProcessor();
    }

    @Bean
    public ItemWriter<Transaction> itemWriter(Marshaller marshaller) {
        StaxEventItemWriter<Transaction> itemWriter = new StaxEventItemWriter<>();
        itemWriter.setMarshaller(marshaller);
        itemWriter.setRootTagName("transactionRecord");
        itemWriter.setResource(outputXml);

        return itemWriter;
    }



    @Bean
    public Marshaller marshaller() {
        Jaxb2Marshaller marshaller = new Jaxb2Marshaller();
        marshaller.setClassesToBeBound(Transaction.class);

        return marshaller;
    }


    @Bean
    public Step retryStep(@Qualifier("retryItemProcessor") ItemProcessor<Transaction, Transaction> processor,
      ItemWriter<Transaction> writer) throws ParseException {

        return stepBuilderFactory.get("retryStep")
          .<Transaction, Transaction>chunk(10)
          .reader(itemReader(inputCsv))
          .processor(processor)
          .writer(writer)
          .faultTolerant()
          .retryLimit(3)
          .retry(ConnectTimeoutException.class)
          .retry(DeadlockLoserDataAccessException.class)
          .build();
    }



    @Bean(name = "retryBatchJob")
    public Job retryJob(@Qualifier("retryStep") Step retryStep) {
        return jobBuilderFactory
          .get("retryBatchJob")
          .start(retryStep)
          .build();
    }

这里调用faultTolerant()方法启用重试功能，并设置重试次数和对应异常。

3. 测试重试功能

我们测试场景，期望接口在一定时间内返回年龄和邮编。前两次调用API抛出异常ConnectTimeoutException，第三次成功调用：

@RunWith(SpringRunner.class)
@SpringBatchTest
@EnableAutoConfiguration
@ContextConfiguration(classes = { SpringBatchRetryConfig.class })
public class SpringBatchRetryIntegrationTest {

    private static final String TEST_OUTPUT = "xml/retryOutput.xml";
    private static final String EXPECTED_OUTPUT = "src/test/resources/output/batchRetry/retryOutput.xml";

    @Autowired
    private JobLauncherTestUtils jobLauncherTestUtils;

    @MockBean
    private CloseableHttpClient closeableHttpClient;

    @Mock
    private CloseableHttpResponse httpResponse;

    @Test
    public void whenEndpointAlwaysFail_thenJobFails() throws Exception {
        when(closeableHttpClient.execute(any()))
          .thenThrow(new ConnectTimeoutException("Endpoint is down"));

        JobExecution jobExecution = jobLauncherTestUtils.launchJob(defaultJobParameters());
        JobInstance actualJobInstance = jobExecution.getJobInstance();
        ExitStatus actualJobExitStatus = jobExecution.getExitStatus();

        assertThat(actualJobInstance.getJobName(), is("retryBatchJob"));
        assertThat(actualJobExitStatus.getExitCode(), is("FAILED"));
        assertThat(actualJobExitStatus.getExitDescription(), containsString("org.apache.http.conn.ConnectTimeoutException"));
    }

    @Test
    public void whenEndpointFailsTwicePasses3rdTime_thenSuccess() throws Exception {
        FileSystemResource expectedResult = new FileSystemResource(EXPECTED_OUTPUT);
        FileSystemResource actualResult = new FileSystemResource(TEST_OUTPUT);

        //前两次调用失败，第三次继续执行
        when(httpResponse.getEntity())
          .thenReturn(new StringEntity("{ \"age\":10, \"postCode\":\"430222\" }"));

        when(closeableHttpClient.execute(any()))
          .thenThrow(new ConnectTimeoutException("Timeout count 1"))
          .thenThrow(new ConnectTimeoutException("Timeout count 2"))
          .thenReturn(httpResponse);

        JobExecution jobExecution = jobLauncherTestUtils.launchJob(defaultJobParameters());
        JobInstance actualJobInstance = jobExecution.getJobInstance();
        ExitStatus actualJobExitStatus = jobExecution.getExitStatus();

        assertThat(actualJobInstance.getJobName(), is("retryBatchJob"));
        assertThat(actualJobExitStatus.getExitCode(), is("COMPLETED"));
        AssertFile.assertFileEquals(expectedResult, actualResult);
    }

    private JobParameters defaultJobParameters() {
        JobParametersBuilder paramsBuilder = new JobParametersBuilder();
        paramsBuilder.addString("jobID", String.valueOf(System.currentTimeMillis()));
        return paramsBuilder.toJobParameters();
    }
}

job成功执行，从日志可以看到两次失败，最终调用成功。

19:06:57.758 [main] INFO  o.b.batch.service.RetryItemProcessor - Attempting to process user with id=1234
19:06:57.758 [main] INFO  o.b.batch.service.RetryItemProcessor - Attempting to process user with id=1234
19:06:57.758 [main] INFO  o.b.batch.service.RetryItemProcessor - Attempting to process user with id=1234
19:06:57.758 [main] INFO  o.b.batch.service.RetryItemProcessor - Attempting to process user with id=9999
19:06:57.773 [main] INFO  o.s.batch.core.step.AbstractStep - Step: [retryStep] executed in 31ms

同时也定义了另一个测试，重试多次并失败，抛出异常 ConnectTimeoutException。

4. 总结

本文我们学习如何配置Spring Batch的重试逻辑。通过示例学习配置并机型测试，希望对你有帮助。

本文地址：https://blog.csdn.net/neweastsun/article/details/109264179